Culminating Project Title
Puredata Systems for Analytics: Concurrency and Workload Management
Date of Award
Culminating Project Type
Information Assurance: M.S.
Information Assurance and Information Systems
Herberger School of Business
Dr. Jim Q Chen
Dr. Susantha Herath
Dr. Changsoo Sohn
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
PureDataTM System for Analytics also called as Netezza is a data warehouse server handling analytic operations capable of providing throughput 1000 times greater and faster than traditional database servers. Impressively, it requires minimal system tuning thereby providing high-end performance as well as requiring a low total cost of ownership (TCO). Database performance is directly linked to the allocation of system resources on a database management system. The heart of the Netezza appliance, Field-Programmable Gate Array (FPGA) plays a key role in boosting the overall performance of a server. I/O operations are always a bottleneck in any database server and it is the FPGA that eradicates the I/O problem in Netezza by filtering the data across each snippet processing unit (SPU), processing and running the queries faster thereby pumping server’s performance greatly.
This paper describes the current problems the companies face in a “big data” environment which includes concurrency handling and query performance. There are various factors which affect a query's performance, which includes bad data distribution, stale statistics, server load and uneven system resources. Since this paper is restricted to only the system resources, an in-depth analysis of system resources and its components will be analyzed in this research. A database server’s performance is directly related to its underlying allocation of system resources. Work Load Management (WLM) and each of its features are described in this paper which gives the reader a clear notion of how a query's performance is altered using various mechanisms. The paper describes the current performance problems that exist on the traditional database servers and how the Work Load Management components can be tweaked along with the predefined system configurations to process a query to run faster on a Netezza machine.
Muddu, Sai Mohit, "Puredata Systems for Analytics: Concurrency and Workload Management" (2015). Culminating Projects in Information Assurance. 3.